The synaptic mechanisms that underlie the brain's remarkable capacity for learning remain unclear. One new model predicts that the acquisition and storage of learned skills involves the experience-driven clustering of co-active synapses onto short lengths of dendrite. Direct evidence for this synaptic clustering hypothesis is scant. This research proposal describes the first comprehensive tests of the model, to be provided by quantitative analysis of microanatomical changes in the barn owl auditory localization pathway. We recently showed that auditory-visual learning in juvenile owls involves selective adjustments in synaptic clustering on the dendrites of space-specific neurons, located within the inferior colliculus. In the first two sets of experiments we will determine whether similar changes occur in adults. Using multiple paradigms we will examine changes related both to the re-expression of normal circuitry and the acquisition of novel circuitry. The results will reveal the extent to which normal, enhanced and incremental training paradigms engage the synaptic clustering mechanism. In the third set of experiments we will examine a separate prediction of the model, whether non co-active synapses segregate on dendrites. The final set of experiments is designed to provide validation of these results at 'anatomical ground truth.' Specifically, we will use a newly developed high throughput method for serial section transmission electron microscopy to reconstruct the complete wiring diagram of the normal and learned circuits. These novel connectomes will be useful both for addressing the hypothesis-driven aims of this grant and for discovery-based approaches. Collectively, our experiments promise to reveal basic principles of synaptic reorganization that underlie the normal development and plasticity of sensory circuits, in both juvenile and adult brains. Such knowledge should prove invaluable for the long-term development of therapies aimed at the remediation of learning disorders.